Geographic and racial disparities in stroke mortality have existed for over 50 years with an incremental annual cost of $4.6 billion. Many causes for these disparities have been proposed, but existing research has inadequately addressed these, and interventions have failed to reduce their magnitude. Further, national data are not available describing the prevalence of cognitive decline, a problem largely caused by cerebrovascular disease and its risk factors. With completion of recruitment, baseline evaluation and follow-up of a national cohort of 30,000 black and white community dwelling participants, the REasons for Geographic And Racial Differences in Stroke (REGARDS) study is uniquely positioned to test older but unsubstantiated hypotheses explaining race/region disparities, to assess new evidence-based hypotheses, and to provide the first national data on predictors of cognitive decline. In this continuation application we propose to continue follow-up for incident stroke and cognitive decline among this cohort, which will quadruple the number of observed stroke events arid provide longitudinal data on cognitive decline. These data will provide the opportunity to investigate the underlying causes of the disparities in stroke mortality by assessing the role of confounding factors to mediate the observed excess risk among African Americans and Southerners, and to develop risk functions for relatively rare stroke subtypes including ICH and SAH. Preliminary analysis from REGARDS suggests that there is a-"cognitive decline belt" that is concordant with the Stroke belt. The data from the continuation will allow a refinement of the description of these disparities, and, using similar analysis approaches employed for the stroke outcomes, will provide insights into the causes for these disparities. Finally, we will investigate the role of cognitive decline as a predictor for subsequent stroke events. Completion of these aims will allow for development of interventions to eliminate disparities in the incidence of stroke and cognitive decline. Implementation of specific interventions customized for black and white populations, and by region of the United States, may reduce health disparities in stroke recognition and treatment, risk factor prevention, and optimal management of existing risk factors.